micromachines

Perspective Engineering Microfluidic Organoid-on-a-Chip Platforms

Fang Yu 1, Walter Hunziker 2,3 and Deepak Choudhury 1,* 1 Bio-Manufacturing Programme, Singapore Institute of Manufacturing Technology (SIMTech), Agency for Science, Technology and Research (A*STAR), 2 Fusionopolis Way, #08-04, Innovis, Singapore 138634, Singapore; [email protected] 2 Institute of Molecular and Biology (IMCB), Agency for Science, Technology and Research (A*STAR), 61 Biopolis Drive, Proteos, Singapore 138673, Singapore; [email protected] 3 Department of Physiology, 2 Medical Drive, MD9, National University of Singapore, Singapore 117593, Singapore * Correspondence: [email protected]

 Received: 14 January 2019; Accepted: 22 February 2019; Published: 27 February 2019 

Abstract: In vitro cell culture models are emerging as promising tools to understand human development, disease progression, and provide reliable, rapid and cost-effective results for drug discovery and screening. In recent years, an increasing number of in vitro models with complex organization and controlled microenvironment have been developed to mimic the in vivo structure and function. The invention of organoids, self-organized organ-like cell aggregates that originate from multipotent stem cells, has allowed a whole new level of biomimicry to be achieved. Microfluidic organoid-on-a-chip platforms can facilitate better nutrient and gas exchange and recapitulate 3D architecture and physiology. They have the potential to transform the landscape of drug development and testing. In this review, we discuss the challenges in the current organoid models and describe the recent progress in the field of organoid-on-a-chip.

Keywords: organoids; microfluidics; organ-on-a-chip; microbioreactor; drug screening; cell culture

1. Introduction Organoids are a new type of 3D culture models that have emerged in recent years. They are essentially miniaturized organs generated from stem cells in vitro. Under appropriate growth factor treatments, these cells differentiate and self-organize into organ-specific cell types and tissue organization. This allows recapitulation of the structure and function of the organs [1] and is useful for the study of human and pathology. Organoid models have the potential to complement or even substitute for animal models. Compared with animal models, the organoid model can be scaled-up for high throughput testing at a lower cost with fewer ethical concerns. In addition, it can be used to study human diseases that are difficult to be modeled accurately in animals [2]. Organoids could be produced from various sources such as primary cells [3], pluripotent stem cells [4,5], embryonic [6] or adult stem cells [7], and patient-derived induced pluripotent stem cells [4,8]. Currently, there are three general approaches to generate organoid in vitro: (1) organoid formation on extracellular matrix (for example ) scaffolds [3]. As an example for this approach, a single Lgr5+ present at the tip of the intestinal crypt could self-organize into a crypt-villi structure and form intestinal an organoid without the support of mesenchymal tissue. (2) Organoid formation from (EB) by agitation using spinning bioreactors [4]. This approach is often used to produce cerebral and retinal organoids. As demonstrated by Lancaster et al., skin fibroblast cells from a patient with microcephaly are reprogrammed using lentiviral delivery of the four major reprogramming factors, Oct4, Sox2, Klf4, and c-Myc. These induced pluripotent stem cells (iPSCs) form EBs, as an intermediate stage and develop to organoids [4]. (3) Air-liquid interface (ALI) method [9,10]. In this

Micromachines 2019, 10, 165; doi:10.3390/mi10030165 www.mdpi.com/journal/micromachines Micromachines 2019, 10, x 2 of 13

of the four major reprogramming factors, Oct4, Sox2, Klf4, and c‐Myc. These induced pluripotent Micromachines 2019, 10, 165 2 of 12 stem cells (iPSCs) form EBs, as an intermediate stage and develop to organoids [4]. (3) Air‐liquid interface (ALI) method [9,10]. In this approach, the top layer of the cells is exposed to air, and the approach,basal surface the is top in layer contact of thewith cells the is liquid exposed medium to air, i.e., and the the organoids basal surface are cultured is in contact in a with gel matrix the liquid and mediumtheir lumen i.e., is the directly organoids exposed are cultured to air instead in a gel of matrixsubmerged and theirin culture lumen media. is directly The exposedALI method to air is insteadcurrently of mainly submerged used in to culture produce media. kidney The organoids ALI method [10] isand currently intestinal mainly organoids used to[9]. produce kidney organoidsIn the [10 following] and intestinal sections organoids of this [9review,]. we began by assessing the critical limitations of traditionalIn the following in vitro models sections and of thiscurrent review, organoids we began models. by assessing In Section the critical 3, we discussed limitations advantages of traditional of inmicrofluidics vitro models for and cell current culture, organoids and highlighted models. In the Section elements3, we which discussed we advantagesthink would of be microfluidics particularly foruseful cell culture,for organoid and highlighted models. In the Section elements 4, while which discussing we think would various be particularlyorganoids‐on useful‐chip for studies, organoid we models.highlighted In Sectionwhich 4particular, while discussing advantage/element various organoids-on-chip of microfluidics has studies, played we a crucial highlighted role to which make particularthe model advantage/element successful. In the oflast microfluidics section of the has manuscript, played a crucial we summarized role to make the the future model challenges successful. in Inreproducibility, the last section ofscaling the manuscript,‐up, vascularization, we summarized maturation, the future challengespharmacokinetics/pharmacodynamics in reproducibility, scaling-up, vascularization,modeling, and getting maturation, clinical pharmacokinetics/pharmacodynamics viable organoid models. modeling, and getting clinical viable organoid models. 2. Limitations of Current Organoid Models 2. Limitations of Current Organoid Models In this section, we highlight some of the main limitations for traditional in vitro culture models and Inorganoid this section, models; we these highlight limitations some of need the to main be addressed limitations when for traditional organoidsin are vitro usedculture for advanced models anddevelopmental organoid models; biology these studies limitations and drug need screening to be addressed applications when (Figure organoids 1). are used for advanced developmental biology studies and drug screening applications (Figure1).

Figure 1. Limitations and goal of current organoid models. (a) Traditional in vitro models are too simplified, complex organoid models with multiple cell types and 3D architecture can be developed toFigure better 1. recapitulate Limitations inand vivo goalorgans. of current (b) There organoid is a lackmodels. of nutrient (a) Traditional exchange in atvitro the models interior are of thetoo organoid,simplified, introducing complex organoid flow and models improving with multiple nutrient andcell types gas exchange and 3D architecture will help to createcan be largerdeveloped and moreto better mature recapitulate organoids. in (vivoc) Current organs. organoid (b) There technology is a lack hasof nutrient limited uniformityexchange at and the reproducibility, interior of the withorganoid, better introducing geometrical flow confinement and improving and environmental nutrient and control, gas exchange future organoidswill help to production create larger will and be moremore reproducible. mature organoids. (c) Current organoid technology has limited uniformity and reproducibility, with better geometrical confinement and environmental control, future organoids production will be Problemmore reproducible. 1: Traditional in vitro culture models are too simplified to represent complex 3D tissues with multiple cell types (Figure1a). This is due to the limited number of cell types and simplified environmentalProblem 1 cues: Traditional in the 2D in models. vitro culture Conventionally, models are cells too aresimplified cultured to withrepresent 2D tissue complex culture 3D techniquestissues within multiple vitro. These cell simplifiedtypes (Figure culture 1a). system This is has due provided to the limited important number and of relevant cell types insights and intosimplified . environmental However, cues in animal in the tissues 2D models. and 3D Conventionally, cell culture models, cells are cells cultured and their with surrounding 2D tissue microenvironmentculture techniques interact in vitro. in These all three simplified dimensions culture [11] system and enable has usprovided to have important a better understanding and relevant ofinsights cellular into behavior, cell biology. both in However, vitro and inin vivo animal[12]. tissues When culturedand 3D cell as 3D culture models, models, cells exhibit cells and features their thatsurrounding are highly microenvironment similar to the complex interactin vivo in allconditions three dimensions and show [11] significant and enable improvement us to have in a termsbetter ofunderstanding cell count, cell of morphology, cellular behavior, cell proliferation, both in vitro and in cell vivo differentiation [12]. When cultured [13]. Due as to 3D the models, lack of cells 3D environmentalexhibit features cues, that certain are highly research similar questions, to the such complex as those in vivo relating conditions to the structural and show development significant

Micromachines 2019, 10, 165 3 of 12 of neural tissues as well as neuro-degeneration, cannot be modeled and crucial questions not being properly answered using simplified 2D cell culture system [14]. Spheroids are one of the most commonly used in vitro 3D tissue culture models. They are formed by the aggregation of the cell and are often used in long-term culture [15]. Upon aggregating into spheroids, cells can establish contacts and create a microenvironment that allows the expression of tissue-like phenotypes [16]. However, most spheroid culture models contain only one cell type and do not completely capture the complex intercellular interaction between different cell types [17]. Compared with spheroids, organoids develop from stem cells or organ progenitors and self-organizes in a manner similar to in vivo. However, many current organoid models also do not take into account of blood cells or shear stress by blood flow, stroma, and immune cells [18]. Tumors found in vivo, for example, display complex genetic heterogeneity. On the other hand, it is still not clear whether the tumor organoids created in vitro recapitulates the original tumors and whether it harbors all the complexity as seen in the tumors in vivo [19]. Tumors in vivo often depend on the vascular network to deliver nutrients and oxygen, the angiogenesis process is stimulated by a variety of angiogenic mediators [20], this is not easy to recreate with existing in vitro models. In existing 2D and 3D models, for example, there is often lack of microglia, the resident macrophages of the brain. Microglia are non-ectoderm (e.g., non-neuronal)-derived and are generally absent from the classic 2D cell culture systems or the more advanced 3D cell culture systems that go through a neuroectoderm intermediate. Importantly, microglia have been shown to be an important component in neurological diseases [21] and as a consequence brain organoid models without microglia lack a neuroinflammatory component that is essential for pathological events observed in human Alzheimer’s disease (AD) patients and AD mouse models [22]. Introducing microglia to the culture system will be beneficial if they are able to migrate in and take up residence as functional and ramified microglia. Problem 2: Lack of nutrient/waste/gaseous exchange (Figure1b)—One of the biggest obstacles in growing mature brain organoids is the restricted nutrient supply, gas exchange and waste removal at the interior of the organoids. The average diameter of organoids achieved in most studies is usually up to 3 mm [23,24] It is challenging to generate brain organoids in a biomimetic microenvironment favorable for brain development. Current brain organoid technologies are not capable of generating brain organoids that mature beyond the prenatal brain equivalent. In the process of brain organoid formation process, the EBs are initially encapsulated into Matrigel for culture [25]. Then, they are transferred into Petri dishes or spinning bioreactors for suspended culture to generate brain organoids. However, the major issue with the EBs is the lack of vascularization (even with the use of spinning bioreactor) which still limits the growth and maturation of organoids [26]. The brain organoid produced using this method mimics the initial human brain development but fails to show tissue maturation and complexity as seen in the adult human brain [4]. For many disorders and diseases that develop after the fetal development stage, the EB model is limited. Introducing perfusion flow through endothelial cells-lined channels to the organoid could prevent necrosis at the core of the organoids and enable the generation of a larger-sized organoid that develops past the prenatal stage type [27]. More complex organoids containing endothelial cells and blood vessel-like structures can also be created. Problem 3: Lack of standardized organoids (Figure1c)—Current organoid technology has limited uniformity and reproducibility, making it difficult to be used for toxicity screening or high-throughput testing [4,28]. This is due to inadequate engineering of the cellular microenvironment and the extra-cellular matrix (ECM). To use the organoids in drug screening or another medium to high-throughput applications, the organoids must be generated in a reproducible way [4]. Consistencies in the size and shape of the organoids will allow drug responses to be quantified more accurately. Organoids are conventionally generated in well plates and Petri dishes under static conditions [26]. Without any physical constraints or culture scaffolds, there is very limited control over the size and geometry of the organoid. The differences in the size, shape and cell numbers and their relative arrangement within each organoid leads to difficulties in normalizing the pharmacokinetics profile of drug candidates. Micromachines 2019, 10, 165 4 of 12

3. Developing Advanced Microfluidic Platforms to Improve Current Organoid Models In 3D organoid models, as organoids increase in size and volume, the core becomes distant from the surface in contact with the fresh medium. Simple diffusion process provides insufficient oxygen and nutrient to the growing cells and limits the amount of waste being removed from cells in the core. Consequently, only cells in contact with fresh medium survive [29]. With microfluidic technology, tissue culture can be carried out in a controlled environment that optimizes temperature, pH, nutrient and oxygen supply and waste removal) [30,31]. Advances in microfluidic technology allow us to engineer the organoids with essential structural and physiological features in a controlled manner [11,32] and provide microscale structures and parameters that approximate the conditions in vivo [33]. Moreover, sensors and actuators can be integrated with the microfluidic devices to enable precise monitoring and control [34]. Optimization of the key parameters, such as the cell-cell and cell-ECM contact, cell type composition, tissue architecture, nutrient exchange, and various physical and electrical stimulation, may greatly minimize batch-to-batch variations and increasing fidelity. When accompanied by microfluidic technology, 3D cell culture can be enhanced to become more complex organ-on-a-chip and organoid-on-a-chip models [31,35,36]. Microfluidic organs-on-a-chip platforms have been recently developed to create a variety of biomimetic organ models, such as lung [37,38], [31,39], kidney [40,41], heart [42,43] and neural networks [44]. They also allow integration of multiple tissue compartments to simulate multiple organs, and make systematic pharmacokinetics predictions for new drugs. Multi-organ chips have been made to capture the physiological complexity in the human body and study interaction between different organs [45–47]. It is also possible to use microfluidic approaches or substrates with different stiffness. These will subject cells to different forms of mechanical and physiological stress. Microfluidic platforms are often combined with flexible cell culture scaffold such as hydrogels. They are commonly used to create an in-vivo-like microenvironment for cell culture. They provide functional support for the cells–functioning as an extracellular matrix (ECM) while promoting survival, proliferation, and differentiation [48]. This is achieved by the natural property of hydrogels, which have interconnected pores with high water retention that allows nutrients to be transported to the cells efficiently. ECMs provide structural and biochemical support to the cells and promote cell-ECM interaction and growth [49,50]. Cell-ECM and cell-cell interaction can be tuned by changing the mechanical properties (for example stiffness) and compositions of each material in the scaffold [51]. Microfluidic approaches have advantages over other in vitro culture models since they offer better control over the physical and chemical parameters, the design of complex structures and the use of multiple materials to better mimic the in vivo organs. We summarize the advantages and disadvantages of each model in Table1.

Table 1. Comparison of advantages and disadvantages of microfluidic chips and other in vitro culture models.

In Vitro Culture Models Advantages Disadvantages Static condition Lack of physical and biochemical 2D cell culture Well established protocol chemical cues (culture dish, transwell membrane, Easy to handle and quantify Large media volume and culture flask) Large variation in nutrient and waste concentration 3D cell culture Include cell-cell and cell-ECM interaction Static condition (engineered culture scaffold, Capture the 3D architecture of tissue culture Inefficient nutrient and waste spheroid, microcarrier, tissue Sensitive to drug treatment transport biopsy, organoid) Fine control over microenvironment Difficult to standardize and scale up Microfluidic chip Good mass transport provided by fluid flow Require external pumps, tubing, (Organ-on-a-chip) Ability to integrate with various sensors connectors, and valve to operate and actuators Micromachines 2019, 10, 165 5 of 12

4. Organoid-on-a-Chip Models Microfluidic organs-on-a-chip is an emerging 3D cell culture field that recreate the structural and functional features of human organs [36,52–54]. Organ-on-a-chip models have controlled fluid flow, cell-cell interaction, matrix property, as well as biochemical and biomechanical cues [37,55,56]. The organ-on-a-chip approaches enable development of organoids to millimeter diameter with enhanced nutrient exchange to prevent cell death at the core of the organoids. Karzbrun et al. for example studied the wrinkling and folding mechanism during brain development process with the help of microfabricated organoid-on-a-chip [57] (Figure2a). They observed the appearance of convolutions after the organoid reached a certain cell density and nuclear strain. With the help of in situ imaging, they identified two opposing forces responsible for the differential growth in the organoid that leads to surface wrinkling: cytoskeleton contraction in the organoid core and nuclear expansion at the organoid perimeter. In the organoid-on-a-chip device, human embryonic stem cells (ESCs) were cultured inside the Matrigel-filled compartment, where they undergo self-organization into a shell-like structure with a lumen. In the course of weeks, the living organoids are sandwiched between a porous membrane and a glass coverslip, allowing nutrient to be exchanged and fluorescent live imaging to be carried out. They examined the nuclear motion and swelling during the cell cycle and quantitatively modeled the physics of the folding brain by using high-resolution imaging of single cells and image analysis (Figure2b). In an experiment to treat the organoids acutely with blebbistatin, the actin cytoskeleton in the cells was disturbed. With this experiment, they concluded that cytoskeletal forces could maintain the organoid core contraction and provide the intracellular mechanical stiffness. They carried out a study to model a severe smooth brain malformation called lissencephaly in vitro. To induce the condition in the organoid, the LIS1 gene was edited with CRISPER/Cas9 genome editing. Modified ECM and cytoskeleton formation and reduced cell elasticity were observed in the LIS1 mutant organoids. In this study, the use of a microfabricated-chip allowed the organoid to be supported by a Matrigel scaffold while being physically confined in a small compartment to allow wrinkling and folding to initiate. At the same time, diffusion of nutrient and waste could occur through the porous membrane. This is otherwise unachievable in the conventional organoid model. Overall, the organoid-on-a-chip was demonstrated to be a useful model to study the biological and biophysical mechanism of brain development. Another microfluidic brain organoid-on-a-chip model was developed recently by Wang et al., the organoid culture and differentiation were done in situ under perfusion culture [58] (Figure2c). Compared with other organoid generation methods, where cells from EB differentiate into organoid in 3D suspended culture [4,59], this organoid-on-a-chip creates a platform that allows the EB to mature into self-organized organoids under a controlled microenvironment. The system recapitulates the in vivo organ by incorporating 3D culture Matrigel scaffold, dynamic perfusion flow, and multicellular heterogeneous tissue architecture. Similar to the conventional organoid generation method, organoids cultured on this microfluidic platform exhibited well-defined neural differentiation, regionalization, and cortical organization. Compared with 2D-cultured organoids, the perfusion-cultured organoids expressed enhanced levels of cortical markers (TBR1 and CTIP2), indicating differentiation and progression of . In a follow-up study by the same group [60] (Figure2d), the same microfluidic brain organoid platform mentioned above was used to examine the effect of prenatal nicotine exposure on brain development. The organoids exhibited typical features of a fetal brain during development. When subjected to nicotine exposure, premature differentiation and abnormal brain regional growth were observed. The chip allowed in situ tracking and imaging of the organoid, enabling the visualization of the brain regionalization process as well as the monitoring on the chip the nicotine exposure-induced premature neural differentiation by immunohistochemical staining. The field of organ-on-a-chip has been around for almost 10 years [37]. Although a variety of microfluidic organ-on-a-chip devices with biomimetic compositions, architecture, and functions have been created, there are limited efforts in using reporters and microsensors to monitor different cellular Micromachines 2019, 10, 165 6 of 12 parameters such as growth or metabolism and changes in the culture in the system. This is especially important for drug testing studies in organ-on-a-chip devices which requires an extended period of culture time. To bridge this gap, Zhang et al. developed an integrated modular platform with fluidics-routing breadboard as well as physical, biochemical and optical sensing capabilities [61] (Figure2e). They demonstrated the generation of liver and heart organoids on this platform and validated the capability of the in situ biosensors. The whole platform was encased inside a customized benchtop incubator, including the automatic pneumatic valves for fluidic control, the electronics for physical sensors, and a potentiostat controlled by a centrally programmed computer. The microfluidic device consists of a number of modules: microbioreactors, breadboards, reservoir, a bubble trap, physical sensors, and electrochemical sensors. The label-free immunobiochemical sensors monitor the soluble biomarkers secreted by organoids (Figure2f); the detection is based on the change of interfacial 4−/3− electron transfer kinetics of redox probe [Fe(CN)6] upon antibody-antigen binding. Upon saturation of captured antigen binding, they adapted a mechanism to regenerate the electrode surface so it can be reused for subsequent measurements. The immunobiosensing mechanism developed here is a promising tool because by using different antibodies, the functionalized biosensors were able to measure a wide range of soluble biomarkers secreted by the organoids. Apart from the biochemical sensor, a physical sensing unit to monitor temperature, pH, and oxygen levels were installed as well (Figure2f). Inside the microbioreactor, the organoids were constructed using a micropatterning technique through a photo mask. Cells were then encapsulated in gelatin methacryloyl (GelMA) and loaded in the microbioreactor to form spheroid in situ. Finally, they demonstrated that the system could perform automated sensing of drug-induced organoid toxicity using the biochemical sensors at various drug concentrations. This work showcases efforts towards the adaptation of organoids-on-a-chip for in situ organoid generation and automated drug testing. The intestine organoid culture models derived from Lgr5+ intestinal stem cells are usually embedded within Matrigel [3]. They can be propagated indefinitely under growth factor treatment. The organoid model recapitulates the heterogeneity of the intestinal epithelial layer and is suitable for in situ imaging of intestinal development. However, the presence of an enclosed organoid lumen is non-physiological, since secreted biochemical and waste products concentrate within the central lumen instead of being transported through peristalsis and luminal flow. Moreover, as mentioned in the previous section, organoid cultures lack tissue-tissue interfaces, mechanical stimuli, and interaction with other cell types such as immune cells. To this end, Kasendra et al. developed a more complex and physiologically relevant microfluidic intestinal culture system using human cells derived from organoids generated from biopsies of the human intestine [62]. This microfluidic chip enables 3D intestinal villi-like structures to be formed in situ. This chip contains two culture channels: an epithelial channel and a vascular channel. They are separated by a thin (50 µm) flexible ECM-coated porous PDMS membrane containing 7 µm diameter pores (Figure2g). The epithelial and vascular channels are flanked on either side by two hollow vacuum chambers that permit application of cyclic suction to mechanically stretch and relax the microchannels to emulate the peristaltic motions of a living human small intestine. The epithelial cells cultured on the chip exhibited multi-lineage differentiation, established epithelial barrier function, and produced mucus. The transcriptome analysis of the cells cultured on the intestine chip more closely resembled that of adult human duodenum in vivo than the original organoids that were used to generate the cells to seed in the chips. The study exemplifies an improvement from past in vitro intestine microfluidic models [63,64] by incorporating primary cells isolated from patient-specific biopsy-derived organoids as well as gut microvascular endothelial cells. The model supports the formation of villus-like structures without the need to use a micropatterned villus-shaped flexible scaffold. Microfluidic chips can also be used as a simple and robust strategy to generate organoids. Wang et al. fabricated a micropillar chip perfusion system to generate liver organoids from EBs. The hiPSCs went through hepatic differentiation and maturation in situ (Figure2h) [ 58]. A sizable number of human hepatic organoids can be generated this way in a straightforward and controlled Micromachines 2019, 10, 165 7 of 12 way. To control the size and uniformity of the EBs and organoids, the dimensions of the micropillar array were optimized. This micropillar chip array method allowed the formation of uniform EBs and generation of liver organoids afterward in a confined space, avoiding the tedious procedures in the conventional organoid generation methods. The media perfusion in the culture chamber facilitated nutrient transport and waste removal. This study is an excellent example of how engineering approaches using microfluidic chips can address the issue of lack of standardization in conventional Micromachinesorganoid generation 2019, 10, x processes. 8 of 13

Figure 2. Cont.

MicromachinesMicromachines 20192019,,10 10,, 165 x 98 of 1213

Figure 2. Current organoid-on-a-chip models. (a) Brain organoid development in the microchip Figure 2. Current organoid‐on‐a‐chip models. (a) Brain organoid development in the microchip compartment. The top membrane is coupled to a media reservoir, and the bottom coverslip enables compartment. The top membrane is coupled to a media reservoir, and the bottom coverslip enables in situ imaging. Wrinkling is caused by nuclear motion and position-dependent nuclear swelling. Thein situ differential imaging. growth Wrinkling leads is to caused residual by stressnuclear and motion wrinkling and inposition the organoid‐dependent [57]. nuclear (b) Fluorescence swelling. imagesThe differential showing thegrowth development leads to ofresidual the organoid stress and embedded wrinkling in Matrigel, in the organoid and the emergence [57]. (b) Fluorescence of wrinkles. Arrowsimages indicateshowing initial the development wrinkling instability of the [organoid57]. (c) The embedded configuration in Matrigel, of the brain and organoids-on-a-chip the emergence of devicewrinkles. and Arrows the procedures indicate for initial brain organoidswrinkling generation instability on [57]. the chip.(c) The The configuration EBs formed by of hiPSCs the brain were embeddedorganoids‐ inon Matrigel,‐a‐chip device and the and mixtures the procedures were infused for brain into organoids the culture generation channel. The on EBsthe differentiatedchip. The EBs andformed self-organized by hiPSCs intowere brain embedded organoids in Matrigel, [58]. (d) abnormaland the mixtures neurite outgrowth were infused induced into bythe nicotine culture exposurechannel. The in the EBs brain differentiated organoid-on-a-chip and self‐organized [60]. (e) The into integrated brain organoids microfluidic [58]. ( deviced) abnormal consisting neurite of modularoutgrowth components induced by including nicotine microbioreactors, exposure in the breadboard, brain organoid reservoir,‐on‐a‐ bubblechip [60]. trap, (e physical) The integrated sensors, andmicrofluidic electrochemical device biosensorsconsisting [of61 ].modular (f) Continual components measurements including of microbioreactors, temperature, pH, andbreadboard, oxygen concentrationreservoir, bubble within trap, the integrated physical organoid-on-chips.sensors, and electrochemical The organoid-on-chip biosensors allows [61]. in-line (f) automatedContinual electrochemicalmeasurements measurementsof temperature, of albumin pH, andand GST- oxygenα secreted concentration from the hepatic within organoids the integrated as well as CK-MBorganoid from‐on‐ thechips. cardiac The organoids.organoid‐on Beating‐chip allows analysis in of‐line the automated cardiac organoids electrochemical can also be measurements performed [61 of]. (albuming) Schematic and representationGST‐α secreted of from the intestine the hepatic chip, organoids showing the as epithelialwell as CK (blue)‐MB and from microvascular the cardiac (pink)organoids. microchannels Beating analysis separated of bythe a cardiac porous PDMSorganoids membrane can also sandwiched be performed in-between. [61]. (g) The Schematic elastic membranerepresentation can beof extendedthe intestine and retractedchip, showing by the the application epithelial of cyclic(blue) vacuumand microvascular to the hollow (pink) side chambers.microchannels This actuationseparated causes by a theporous mechanical PDMS deformation membrane ofsandwiched the tissue layers in‐between. cultured The in the elastic chip. (membraneb) Procedure can to be establish extended the microfluidicand retracted co-cultures by the application of the primary of cyclic human vacuum intestinal to the epithelium hollow andside intestinalchambers. microvascular This actuation endothelium causes the mechanical in the intestine deformation chip [62 ].of ( hthe) Generation tissue layers of cultured hiPSC-derived in the chip. liver organoids(b) Procedurein vitro to andestablish the configuration the microfluidic of the co liver‐cultures organoid-on-a-chip of the primary system. human [ 58intestinal]. Reproduced epithelium with permissionand intestinal from microvascular [57,58,60–62 ].endothelium in the intestine chip [62]. (h) Generation of hiPSC‐derived liver organoids in vitro and the configuration of the liver organoid‐on‐a‐chip system. [58]. Reproduced with permission from [57,58,60–62].

Micromachines 2019, 10, 165 9 of 12

5. Conclusions and Future Perspective Organoids have helped to expand our understandings of human organ development and have become a useful tool to model human diseases and test therapeutics in vitro. In addition, they contribute to reducing the use of animal models and cost in the pharmaceutical industry. Despite recent developments in the field of organoid-on-a-chip, the full potential of the microfluidic organoid platforms is yet to be realized but will further improve with more refined technical innovations. Limitations of current organoids, which are still highly simplified models of human tissues or organs, may be improved by incorporating microfluidic platforms. The lack of nutrient and gas exchange hinders the maximum size and extent of tissue maturation of the organoids. Exposure of cells to physiological shear flow, mechanical stress, and substrate stiffness can have profound effects on cell and tissue physiology. There is still limited control over the size, shape and relative arrangement of different cell types within the organoids in 3D format, limiting their applications for reproducible quantitative studies as required for robust drug screening and testing. We envision that the next generation of organoid-on-a-chip will have the potential to be scaled-up to be suitable for high-throughput analysis and commercial applications. This will require the generation of reproducible organoids with respect to cell-type composition and arrangement, and overall organoid structure and organization. To increase the critical size of the organoids, more efforts will be made towards promoting vascularization of organoids, in the form of artificial blood vessels in the organoids or spontaneously generated vessels due to activation of angiogenic pathways in incorporated vascular cell types. In order to incorporate multiple organoid types into single chips, the culture medium and physical conditions in the microfluidic platform may need to be optimized for each organoid type. The ultimate goal for microfluidic multi-organoid systems is to achieve in vivo-like settings that capture the structure and physiology of the different organ systems and recapitulate inter-organ interactions and crosstalk. Future engineering approaches will enable organoid models to be utilized in clinical trials and elucidate personalized pathology mechanism and drug responses. To realize this, the time needed to generate fully functional organoids need to be shortened. Currently, it can take months to build an organoid from stem cells derived from primary tissue, this is too long for clinicians to make effective decisions. Due to the difference in systemic drug distribution profile and immune responses that are not observed in vitro, in vivo drug responses are often different from those determined in vitro. More extensive engineering and screening effort are needed to provide proper information about physiologically relevant pharmacokinetic/pharmacodynamics (PK/PD) profile. We believe that appropriate clinical tissue model can be created by combining the ability of microfluidic platforms to precisely control drug input and distribution, with that of organoid models to recapitulate tissue organization and function.

Author Contributions: F.Y.: Design, manuscript writing including the figures and tables; W.H.: manuscript writing and editing; D.C.: Conception and design, manuscript editing; F.Y., W.H. and D.C.: final approval of the manuscript. Acknowledgments: This work was supported by A*STAR JCO Grant 15302FG152 awarded to D.C. Conflicts of Interest: The authors indicated no potential conflicts of interest.

References

1. Van der Helm, M.W.; van der Meer, A.D.; Eijkel, J.C.; van den Berg, A.; Segerink, L.I. Microfluidic organ-on-chip technology for blood-brain barrier research. Tissue Barriers 2016, 4, e1142493. [CrossRef] [PubMed] 2. Fatehullah, A.; Tan, S.H.; Barker, N. Organoids as an in vitro model of human development and disease. Nat. Cell Biol. 2016, 18, 246. [CrossRef][PubMed] Micromachines 2019, 10, 165 10 of 12

3. Sato, T.; Vries, R.G.; Snippert, H.J.; Van De Wetering, M.; Barker, N.; Stange, D.E.; Van Es, J.H.; Abo, A.; Kujala, P.; Peters, P.J.; et al. Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature 2009, 459, 262–265. [CrossRef][PubMed] 4. Lancaster, M.A.; Renner, M.; Martin, C.A.; Wenzel, D.; Bicknell, L.S.; Hurles, M.E.; Homfray, T.; Penninger, J.M.; Jackson, A.P.; Knoblich, J.A. Cerebral organoids model human brain development and microcephaly. Nature 2013, 501, 373–379. [CrossRef][PubMed] 5. Quadrato, G.; Nguyen, T.; Macosko, E.Z.; Sherwood, J.L.; Yang, S.M.; Berger, D.R.; Maria, N.; Scholvin, J.; Goldman, M.; Kinney, J.P.; et al. Cell diversity and network dynamics in photosensitive human brain organoids. Nature 2017, 545, 48–53. [CrossRef][PubMed] 6. Völkner, M.; Zschätzsch, M.; Rostovskaya, M.; Overall, R.W.; Busskamp, V.; Anastassiadis, K.; Karl, M.O. Retinal Organoids from Pluripotent Stem Cells Efficiently Recapitulate Retinogenesis. Stem Cell Rep. 2016, 6, 525–538. [CrossRef][PubMed] 7. Broutier, L.; Andersson-Rolf, A.; Hindley, C.J.; Boj, S.F.; Clevers, H.; Koo, B.K.; Huch, M. Culture and establishment of self-renewing human and mouse adult liver and pancreas 3D organoids and their genetic manipulation. Nat. Protoc. 2016, 11, 1724–1743. [CrossRef][PubMed] 8. Crespo, M.; Vilar, E.; Tsai, S.Y.; Chang, K.; Amin, S.; Srinivasan, T.; Zhang, T.; Pipalia, N.H.; Chen, H.J.; Witherspoon, M.; et al. Colonic organoids derived from human induced pluripotent stem cells for modeling colorectal and drug testing. Nat. Med. 2017, 23, 878–884. [CrossRef][PubMed] 9. Ootani, A.; Li, X.; Sangiorgi, E.; Ho, Q.T.; Ueno, H.; Toda, S.; Sugihara, H.; Fujimoto, K.; Weissman, I.L.; Capecchi, M.R.; et al. Sustained in vitro intestinal epithelial culture within a Wnt-dependent stem cell niche. Nat. Med. 2009, 15, 701–706. [CrossRef][PubMed] 10. Rossi, G.; Manfrin, A.; Lutolf, M.P. Progress and potential in organoid research. Nat. Rev. Genet. 2018, 19, 671–687. [CrossRef][PubMed] 11. Huh, D.; Hamilton, G.A.; Ingber, D.E. From 3D cell culture to organs-on-chips. Trends Cell Biol. 2011, 21, 745–754. [CrossRef][PubMed] 12. Justice, B.A.; Badr, N.A.; Felder, R.A. 3D cell culture opens new dimensions in cell-based assays. Drug Discov. Today 2009, 14, 102–107. [CrossRef][PubMed] 13. Vinci, M.; Gowan, S.; Boxall, F.; Patterson, L.; Zimmermann, M.; Lomas, C.; Mendiola, M.; Hardisson, D.; Eccles, S.A. Advances in establishment and analysis of three-dimensional tumor spheroid-based functional assays for target validation and drug evaluation. BMC Biol. 2012, 10, 29. [CrossRef][PubMed] 14. Hopkins, A.M.; DeSimone, E.; Chwalek, K.; Kaplan, D.L. 3D in vitro modeling of the central . Prog. Neurobiol. 2015, 125, 1–25. [CrossRef][PubMed] 15. Tong, W.H.; Fang, Y.; Yan, J.; Hong, X.; Singh, N.H.; Wang, S.R.; Nugraha, B.; Xia, L.; Fong, E.L.; Iliescu, C.; et al. Constrained spheroids for prolonged hepatocyte culture. Biomaterials 2016, 80, 106–120. [CrossRef][PubMed] 16. Antoni, D.; Burckel, H.; Josset, E.; Noel, G. Three-dimensional cell culture: A breakthrough in vivo. Int. J. Mol. Sci. 2015, 16, 5517–5527. [CrossRef][PubMed] 17. Mehta, G.; Hsiao, A.Y.; Ingram, M.; Luker, G.D.; Takayama, S. Opportunities and challenges for use of tumor spheroids as models to test drug delivery and efficacy. J. Control. Release 2012, 164, 192–204. [CrossRef] [PubMed] 18. Drost, J.; Clevers, H. Organoids in cancer research. Nat. Rev. Cancer 2018, 18, 7407–7418. [CrossRef][PubMed] 19. Dutta, D.; Heo, I.; Clevers, H. Disease Modeling in Stem Cell-Derived 3D Organoid Systems. Trends Mol. Med. 2017, 23, 393–410. [CrossRef][PubMed] 20. Hillen, F.; Griffioen, A.W. Tumour vascularization: Sprouting angiogenesis and beyond. Cancer Metast. Rev. 2007, 26, 489–26502. [CrossRef][PubMed] 21. Muffat, J.; Li, Y.; Yuan, B.; Mitalipova, M.; Omer, A.; Corcoran, S.; Bakiasi, G.; Tsai, L.H.; Aubourg, P.; Ransohoff, R.M.; et al. Efficient derivation of microglia-like cells from human pluripotent stem cells. Nat. Med. 2016, 22, 1358–1367. [CrossRef][PubMed] 22. Park, J.; Wetzel, I.; Marriott, I.; Dréau, D.; D’Avanzo, C.; Kim, D.Y.; Tanzi, R.E.; Cho, H. A 3D human triculture system modeling neurodegeneration and neuroinflammation in Alzheimer’s disease. Nat. Neurosci. 2018, 21, 941–951. [CrossRef][PubMed] 23. Akkerman, N.; Defize, L.H.K. Dawn of the organoid era: 3D tissue and organ cultures revolutionize the study of development, disease, and regeneration. BioEssays 2017, 39, 1600244. [CrossRef][PubMed] Micromachines 2019, 10, 165 11 of 12

24. Eicher, A.K.; Berns, H.M.; Wells, J.M. Translating Developmental Principles to Generate Human Gastric Organoids. Cell Mol. Gastroenterol. Hepatol. 2018, 5, 353–363. [CrossRef][PubMed] 25. Kurosawa, H. Methods for inducing embryoid body formation: In vitro differentiation system of embryonic stem cells. J. Biosci. Bioeng. 2007, 103, 389–398. [CrossRef][PubMed] 26. Lancaster, M.A.; Knoblich, J.A. Generation of cerebral organoids from human pluripotent stem cells. Nat. Protoc. 2014, 9, 2329–2340. [PubMed] 27. Quadrato, G.; Brown, J.; Arlotta, P. The promises and challenges of human brain organoids as models of neuropsychiatric disease. Nat. Med. 2016, 22, 1220–1228. [CrossRef][PubMed] 28. Kelava, I.; Lancaster, M.A. Stem Cell Models of Human Brain Development. Cell Stem Cell 2016, 18, 736–748. (In English) [CrossRef][PubMed] 29. Sailon, A.M.; Allori, A.C.; Davidson, E.H.; Reformat, D.D.; Allen, R.J.; Warren, S.M. A novel flow-perfusion bioreactor supports 3D dynamic cell culture. J. Biomed. Biotechnol. 2009.[CrossRef][PubMed] 30. Yu, F.; Zhuo, S.; Qu, Y.; Choudhury, D.; Wang, Z.; Iliescu, C.; Yu, H. On chip two-photon metabolic imaging for drug toxicity testing. Biomicrofluidics 2017, 11, 034108. [CrossRef][PubMed] 31. Yu, F.; Deng, R.; Tong, W.H.; Huan, L.; Way, N.C.; IslamBadhan, A.; Iliescu, C.; Yu, H. A perfusion incubator liver chip for 3D cell culture with application on chronic hepatotoxicity testing. Sci. Rep. 2017, 7, 14528. [CrossRef][PubMed] 32. Bhatia, S.N.; Ingber, D.E. Microfluidic organs-on-chips. Nat. Biotechnol. 2014, 32, 760–772. [CrossRef] [PubMed] 33. Choudhury, D.; Mo, X.; Iliescu, C.; Tan, L.L.; Tong, W.H.; Yu, H. Exploitation of physical and chemical constraints for three-dimensional microtissue construction in microfluidics. Biomicrofluidics 2011, 5, 022203. [CrossRef][PubMed] 34. Yin, X.; Mead, B.E.; Safaee, H.; Langer, R.; Karp, J.M.; Levy, O. Engineering Stem Cell Organoids. Cell Stem Cell 2016, 18, 25–38. [CrossRef][PubMed] 35. Li, X.; Valadez, A.V.; Zuo, P.; Nie, Z. Microfluidic 3D cell culture: Potential application for tissue-based bioassays. Bioanalysis 2012, 4, 1509–1525. [CrossRef][PubMed] 36. Yu, F.; Kumar, N.D.S.; Choudhury, D.; Foo, L.C.; Ng, S.H. Microfluidic platforms for modeling biological barriers in the circulatory system. Drug Discov. Today 2018, 23, 815–829. [CrossRef][PubMed] 37. Huh, D.; Matthews, B.D.; Mammoto, A.; Montoya-Zavala, M.; Hsin, H.Y.; Ingber, D.E. Reconstituting organ-level lung functions on a chip. Science 2010, 328, 1662–1668. [CrossRef][PubMed] 38. Benam, K.H.; Villenave, R.; Lucchesi, C.; Varone, A.; Hubeau, C.; Lee, H.H.; Alves, S.E.; Salmon, M.; Ferrante, T.C.; Weaver, J.C.; et al. Small airway-on-a-chip enables analysis of human lung inflammation and drug responses in vitro. Nat. Methods 2016, 13, 151–157. [CrossRef][PubMed] 39. No, D.Y.; Lee, K.H.; Lee, J.; Lee, S.H. 3D liver models on a microplatform: Well-defined culture, engineering of liver tissue and liver-on-a-chip. Lab Chip 2015, 15, 3822–3837. 40. Wang, L.; Tao, T.; Su, W.; Yu, H.; Yu, Y.; Qin, J. A disease model of diabetic nephropathy in a glomerulus- on-a-chip microdevice. Lab Chip 2017, 17, 1749–1760. [CrossRef][PubMed] 41. Jang, K.J.; Mehr, A.P.; Hamilton, G.A.; McPartlin, L.A.; Chung, S.; Suh, K.Y.; Ingber, D.E. Human kidney proximal tubule-on-a-chip for drug transport and nephrotoxicity assessment. Integr. Biol. 2013, 5, 1119–1129. [CrossRef][PubMed] 42. Marsano, A.; Conficconi, C.; Lemme, M.; Occhetta, P.; Gaudiello, E.; Votta, E.; Cerino, G.; Redaelli, A.; Rasponi, M. Beating heart on a chip: A novel microfluidic platform to generate functional 3D cardiac microtissues. Lab Chip 2016, 16, 599–610. [CrossRef][PubMed] 43. Zhang, Y.S.; Arneri, A.; Bersini, S.; Shin, S.R.; Zhu, K.; Goli-Malekabadi, Z.; Aleman, J.; Colosi, C.; Busignani, F.; Dell’Erba, V.; et al. Bioprinting 3D microfibrous scaffolds for engineering endothelialized myocardium and heart-on-a-chip. Biomaterials 2016, 110, 45–59. (In English) [CrossRef][PubMed] 44. Peyrin, J.M.; Deleglise, B.; Saias, L.; Vignes, M.; Gougis, P.; Magnifico, S.; Betuing, S.; Pietri, M.; Caboche, J.; Vanhoutte, P.; et al. Axon diodes for the reconstruction of oriented neuronal networks in microfluidic chambers. Lab Chip 2011, 11, 3663–3673. [CrossRef][PubMed] 45. Lee, H.; Kim, D.S.; Ha, S.K.; Choi, I.; Lee, J.M.; Sung, J.H. A pumpless multi-organ-on-a-chip (MOC) combined with a pharmacokinetic-pharmacodynamic (PK-PD) model. Biotechnol. Bioeng. 2017, 114, 432–443. [CrossRef][PubMed] Micromachines 2019, 10, 165 12 of 12

46. Esch, M.B.; Ueno, H.; Applegate, D.R.; Shuler, M.L. Modular, pumpless body-on-a-chip platform for the co-culture of GI tract epithelium and 3D primary liver tissue. Lab Chip 2016, 16, 2719–2729. [CrossRef] [PubMed] 47. Maschmeyer, I.; Lorenz, A.K.; Schimek, K.; Hasenberg, T.; Ramme, A.P.; Hübner, J.; Lindner, M.; Drewell, C.; Bauer, S.; Thomas, A.; et al. A four-organ-chip for interconnected long-term co-culture of human intestine, liver, skin and kidney equivalents. Lab Chip 2015, 15, 2688–2699. [CrossRef][PubMed] 48. Almany, L.; Seliktar, D. Biosynthetic hydrogel scaffolds made from fibrinogen and polyethylene glycol for 3D cell cultures. Biomaterials 2005, 26, 2467–2477. [CrossRef][PubMed] 49. Pampaloni, F.; Reynaud, E.G.; Stelzer, E.H. The third dimension bridges the gap between cell culture and live tissue. Nat. Rev. Mol. Cell Biol. 2007, 8, 839–845. [CrossRef][PubMed] 50. Choudhury, D.; Tun, H.W.; Wang, T.; Naing, M.W. Organ-Derived Decellularized Extracellular Matrix: A Game Changer for Bioink Manufacturing? Trends Biotechnol. 2018, 36, 787–805. [CrossRef][PubMed] 51. Carletti, E.; Motta, A.; Migliaresi, C. Scaffolds for and 3D cell culture. Methods Mol. Biol. 2011, 695, 17–39. [PubMed] 52. Ghaemmaghami, A.M.; Hancock, M.J.; Harrington, H.; Kaji, H.; Khademhosseini, A. Biomimetic tissues on a chip for drug discovery. Drug Discov. Today 2012, 17, 173–181. [CrossRef][PubMed] 53. Zhang, B.; Korolj, A.; Lai, B.F.L.; Radisic, M. Advances in organ-on-a-chip engineering. Nat. Rev. Mater. 2018, 3, 257–278. [CrossRef] 54. Kimura, H.; Sakai, Y.; Fujii, T. Organ/body-on-a-chip based on microfluidic technology for drug discovery. Drug Metab. Pharmacokinet. 2018, 33, 43–48. [CrossRef][PubMed] 55. Chung, M.; Lee, S.; Lee, B.J.; Son, K.; Jeon, N.L.; Kim, J.H. Wet-AMD on a Chip: Modeling Outer Blood-Retinal Barrier In Vitro. Adv. Healthc. Mater. 2018, 7, 1700028. [CrossRef][PubMed] 56. Schepers, A.; Li, C.; Chhabra, A.; Seney, B.T.; Bhatia, S. Engineering a perfusable 3D human liver platform from iPS cells. Lab Chip 2016, 16, 2644–2653. [CrossRef][PubMed] 57. Karzbrun, E.; Kshirsagar, A.; Cohen, S.R.; Hanna, J.H.; Reiner, O. Human Brain Organoids on a Chip Reveal the Physics of Folding. Nat. Phys. 2018, 14, 515–522. [CrossRef][PubMed] 58. Wang, Y.; Wang, L.; Guo, Y.; Zhu, Y.; Qin, J. Engineering stem cell-derived 3D brain organoids in a perfusable organ-on-a-chip system. RSC Adv. 2018, 8, 1677–1685. [CrossRef] 59. Qian, X.; Nguyen, H.N.; Song, M.M.; Hadiono, C.; Ogden, S.C.; Hammack, C.; Yao, B.; Hamersky, G.R.; Jacob, F.; Zhong, C.; et al. Brain-Region-Specific Organoids Using Mini-bioreactors for Modeling ZIKV Exposure. Cell 2016, 165, 1238–1254. [CrossRef][PubMed] 60. Wang, Y.; Wang, L.; Zhu, Y.; Qin, J. Human brain organoid-on-a-chip to model prenatal nicotine exposure. Lab Chip 2018, 18, 851–860. [CrossRef][PubMed] 61. Zhang, Y.S.; Aleman, J.; Shin, S.R.; Kilic, T.; Kim, D.; Shaegh, S.A.; Massa, S.; Riahi, R.; Chae, S.; Hu, N.; et al. Multisensor-integrated organs-on-chips platform for automated and continual in situ monitoring of organoid behaviors. Proc. Natl. Acad. Sci. USA 2017, 114, E2293–E2302. [CrossRef][PubMed] 62. Kasendra, M.; Tovaglieri, A.; Sontheimer-Phelps, A.; Jalili-Firoozinezhad, S.; Bein, A.; Chalkiadaki, A.; Scholl, W.; Zhang, C.; Rickner, H.; Richmond, C.A.; et al. Development of a primary human Small Intestine-on-a-Chip using biopsy-derived organoids. Sci. Rep. 2018, 8, 2871. [CrossRef][PubMed] 63. Kim, H.J.; Ingber, D.E. Gut-on-a-Chip microenvironment induces human intestinal cells to undergo villus differentiation. Integr. Biol. 2013, 5, 1130–1140. [CrossRef][PubMed] 64. Kim, H.J.; Li, H.; Collins, J.J.; Ingber, D.E. Contributions of microbiome and mechanical deformation to intestinal bacterial overgrowth and inflammation in a human gut-on-a-chip. Proc. Natl. Acad. Sci. USA 2016, 113, E7–E15. [CrossRef][PubMed]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).